Roel Wieringa is Chair of Information Systems at the University of Twente, the Netherlands. His research interests include requirements engineering, IT security risk assessment, and design science research methodology for software. He has written three books, Requirements Engineering: Frameworks for Understanding (Wiley, 1996), Design Methods for Reactive Systems: Yourdon, Statemate and the UML (Morgan Kaufmann, 2003, and Design Science Methodology for Information Systems and Software Engineering (Springer, 2014). He serves on the board of editors of the Requirements Engineering Journal and of the Journal of Software and Systems Modeling. Roel Wieringa has been Associate Editor in Chief of IEEE Software for the area of requirements engineering from 2004 to 2007, has been head of the Computer Science Department of the UT from 2009 to 2013, and has been scientific director of the Dutch national School for Information and Knowledge Systems (SIKS, www.siks.nl), from 2006 to 2011. He currently heads the research group of Services, Cybersecurity, and safety at the UT. Find more information at http://wwwhome.ewi.utwente.nl/~roelw/.
In the past 20 years, empirical software engineering has developed into a discipline of its own, with an international researchers network (ISERN, started in 1993), its own journal (started in 1996) and a dedicated conference (started in 2007). This is strange, for scientific disciplines are usually defined by subject, not by research method. It feeds the misunderstanding that empirical software engineering collects facts rather than produces scientific theories, avoids the creation of useful designs, and imitates the sometimes questionable use of mathematics, in particular of statistics, in the social sciences.
In this talk I will take the point of view that software engineer research is a design science, and that, as all other design sciences, it should include both empirical studies and the design of useful artifacts, and should use mathematics to predict artifact properties. I will review some misconceptions held by designers about empirical research, by empirical researchers about design, and by mathematicians about themselves, or the other way around. In conclusion I will sketch a methodology for software engineering research that avoids these misconceptions and may be useful to raise the defensibility of software engineering research methodology with respect to neighboring disciplines.